# Reducing Compute Waste in LLMs through Kernel-Level DVFS

arXiv: https://arxiv.org/abs/2601.08539 · HTML: https://mlsanity.com/paper/2601.08539/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 30 references that are in the map.
- First posted on arXiv 13 Jan 2026.

## Builds on (most cited first, 9 of 9)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima](https://mlsanity.com/paper/1609.04836/) (2016)
- [Sustainable AI: Environmental Implications, Challenges and Opportunities](https://mlsanity.com/paper/2111.00364/) (2021)
- [DynamoLLM: Designing LLM Inference Clusters for Performance and Energy Efficiency](https://mlsanity.com/paper/2408.00741/) (2024)
- [Part-time Power Measurements: nvidia-smi's Lack of Attention](https://mlsanity.com/paper/2312.02741/) (2023)
- [Predict; Do not React for Enabling Efficient Fine Grain DVFS in GPUs](https://mlsanity.com/paper/2205.00121/) (2022)
- [Going green: optimizing GPUs for energy efficiency through model-steered auto-tuning](https://mlsanity.com/paper/2211.07260/) (2022)
- [Methodology for GPU Frequency Switching Latency Measurement](https://mlsanity.com/paper/2502.20075/) (2025)

## Built on by (most cited first, 2 of 2)

- [KAIROS: Stateful, Context-Aware Power-Efficient Agentic Inference Serving](https://mlsanity.com/paper/2604.16682/) (2026)
- [Energy-Efficient GPU DVFS for Fine-Tuning of SLMs on Resource-constrained Embedded Devices](https://mlsanity.com/paper/2607.05933/) (2026)

## Abstract (arXiv)

The rapid growth of AI has fueled the expansion of accelerator- or GPU-based data centers. However, the rising operational energy consumption has emerged as a critical bottleneck and a major sustainability concern. Dynamic Voltage and Frequency Scaling (DVFS) is a well-known technique used to reduce energy consumption, and thus improve energy-efficiency, since it requires little effort and works with existing hardware. Reducing the energy consumption of training and inference of Large Language Models (LLMs) through DVFS or power capping is feasible: related work has shown energy savings can be significant, but at the cost of significant slowdowns. In this work, we focus on reducing waste in LLM operations: i.e., reducing energy consumption without losing performance. We propose a fine-grained, kernel-level, DVFS approach that explores new frequency configurations, and prove these save more energy than previous, pass- or iteration-level solutions. For example, for a GPT-3 training run, a pass-level approach could reduce energy consumption by 2% (without losing performance), while our kernel-level approach saves as much as 14.6% (with a 0.6% slowdown). We further investigate the effect of data and tensor parallelism, and show our discovered clock frequencies translate well for both. We conclude that kernel-level DVFS is a suitable technique to reduce waste in LLM operations, providing significant energy savings with negligible slow-down.